Summary
What you’ll impact
Our company is seeking a Member of Technical Staff – AI/ML to design, build, and deploy AI-powered systems that solve financial operations challenges for B2B companies. The role focuses on turning cutting‑edge AI research into production‑grade features such as intelligent invoice matching and automated payment reconciliation, working closely with product, engineering, data, and customers.
Responsibilities
What you'll do
- Create production-ready AI applications and agentic systems that address customer financial workflow challenges
- Build tool-using LLM agents that surface insights, recommend next steps, and execute approved tasks — not just chatbots
- Refine capabilities like invoice matching, payment reconciliation, and financial document processing
- Apply and optimize LLMs and RAG systems for financial use cases, including fine-tuning on proprietary data where it moves the needle
- Build robust AI pipelines from ingestion to inference — reliable, maintainable, and cost-efficient through smart model routing
- Stand up golden datasets, agent tracing, regression-on-PR, and A/B testing so we ship confidently and catch silent regressions
- Partner with Product, Engineering, Data, and customers to translate business needs into AI solutions
- Treat every AI feature as a continuously-improving system — instrument everything, iterate
Requirements
What you’ll bring
- Have 5+ years of AI/ML experience
- Have shipped agentic products in production and understand the failure modes (tool use, planning, state, recovery, human-in-the-loop)
- Have integrated and fine-tuned LLMs, and built RAG systems for document- or data-intensive workflows
- Have trained and deployed classical ML models (risk scoring, forecasting, ranking, or similar) — feature engineering, model selection, evaluation, calibration
- Have strong opinions on AI/ML evals — golden datasets, offline + online evaluation, statistical significance, evals in CI
- Are familiar with LLM observability tooling (LangSmith, Braintrust, Arize, or similar) and treat tracing as table stakes
- Understand MLOps fundamentals: deployment, monitoring, A/B testing, model and prompt versioning, feature stores
- Are fluent in Python and modern AI/ML tooling (PyTorch, Transformers, scikit-learn, XGBoost, vLLM, LangChain/LlamaIndex, or equivalent)
- Have shipped AI/ML products that solved real business problems, not just prototypes
- Can translate business requirements into clear technical solutions
- Bonus: experience with model routing across providers
- Bonus: fintech, B2B SaaS, or AR/AP domain experience
- Are plugged into the AI/ML community and energized by bringing AI to real-world use cases